…way" (I changed my mind. We want those zeros. Keeping them allows us to have fewer headers, which can be a bigger win in terms of size of resulting wigs and bigwigs) This reverts commit e789f92.
…side effect of creating lots of headers around zero values
…n wigs. default remains 3
and halPhyloP mask option fix
… functions reading bed streams
…ubtrees can get different steps